Circuit Analysis and Defect Characteristics Estimation Method Using Bimodal Defect-Centric Random Telegraph Noise Model

We propose circuit analysis and defect characteristics estimation methods using the bimodal RTN (random telegraph noise) model of the defect-centric distribution. The bimodal model takes into account the defect characteristics of both high-k and interface layers in gate dielectrics on a 40 nm SiON process, whereas the conventional unimodal model fails to replicate the effects of RTN on the process. The estimation method calculates parameters of the defect characteristics of the bimodal model by using RTN-induced frequency fluctuations measured on the ring oscillators. The analysis method reproduces distributions of RTNinduced frequency fluctuations by the Monte Carlo simulation using the defect parameters. We confirm the proposed methods fully replicates the effect of RTN by comparing simulation and measurement results of a 40 nm test chip. Keywords—RTN (Random Telegraph Noise), defect-centric distribution, variation, reliability, circuit design